People with Down syndrome (DS) are usually in the lowest 2% of the population in intellectual ability, yet their language ability is far lower than their intellectual ability. Severe language limitations have a profound and pervasive impact on life functioning of people with DS. Though much work has been done to understand the pattern of language abilities in individuals with DS, little is known about the nature of the underlying processes or mechanisms that result in this pattern. The broad, long-term objective of the proposed research is to identify cognitive predictors of language impairment in DS, especially syntax. We predict that impairments in implicit learning and phonological memory together lead to impairments in language in this group. The study uses a combined comparative/longitudinal design in which a DS group, age 10-21, and 2 control groups (CA and mixed etiology ID) are tested at Time 1 and the DS and mixed ID groups are tested again 2 years later at Time 2. Specific Aim 1 is to demonstrate that, in addition to the known impairment in phonological memory, there is also a severe impairment in implicit learning in DS. We expect participants with DS to perform more poorly than CA and ID groups on measures of implicit learning and phonological memory. The study also includes a typical developmental trajectory to evaluate implicit learning in the DS and mixed ID groups. Specific Aim 2 is to demonstrate that implicit learning and phonological memory impairments are linked to language impairment in DS. If so, then (a) implicit learning and phonological memory will correlate with concurrent language ability in DS and (b) implicit learning and phonological memory at Time 1 will predict rate of language development over two years in DS. Specific Aim 3 is to test a partial mediation model of the influence of implicit learning on language in DS, especially syntax. We suggest that implicit learning affects language directly, because it limits the acquisition of complex rule systems and covariations necessary for language acquisition, but also indirectly, through phonological memory. Poor implicit learning of covariations among phonologic units in one's native language leads to limitations in phonological memory, which constrain language acquisition. We will test the partial mediation model concurrently and longitudinally over two years. PUBLIC HEALTH RELEVANCE: Individuals with Down syndrome have specific expressive and receptive language difficulties that are even more pronounced than their difficulties in other areas of development. These language delays lead to lifelong communication problems. The focus of this research is to identify memory and learning impairments that are related to the specific pattern of language difficulties that characterize individuals with Down syndrome. Specifically, we propose that a combination of implicit (automatic) learning and phonological memory impairments will be related to the language difficulties experienced by persons with Down syndrome. If supported, these findings will have significant implications for the design of language interventions for this group. They would suggest that language interventions should incorporate more deliberate or explicit techniques to enhance implicit learning processes and support phonological memory during language learning.